Add script to fetch benchmark results for execuTorch (#11734)

# Summary
Provide methods and script to fetch all execuTorch benchamrk data from
HUD API into two dataset,private and public, the script will:
- fetch all data from HUD API from input time range in UTC
- clean out records and tables with only FAILURE_REPORT due to job-level
failures
- get all private table metrics, generate `table_name` and find
intersected public table metrics
- generate private and public table groups
- output data

OutputType:
- run with excel-sheet export
- run with csv export
- run with dataframe format print
- run with json format print

See more guidance in README.md

the data is similar to the excel sheet generated manually in
#10982
The result should be the same as the hud per model datatable:
<img width="1480" alt="image"
src="https://github.com/user-attachments/assets/7c6cc12e-50c5-4ce2-ac87-5cac650486e3"
/>

## helper methods: common.py
provide common.py helper method to convert back csv and excel sheets
back to {"groupInfo":{}, "df":df.DataFrame} format.

# run with
``` bash
python3 .ci/scripts/benchmark_tooling/get_benchmark_analysis_data.py \
--startTime "2025-04-29T09:48:57" \
--endTime "2025-05-13T22:00:00" \
--outputType "excel" \
--models "mv3"

python3 .ci/scripts/benchmark_tooling/analyze_benchmark_stability.py \
--primary-file private.xlsx \
--reference-file public.xlsx
```
Generate excel files:

[private.xlsx](https://github.com/user-attachments/files/20844977/private.xlsx)

[public.xlsx](https://github.com/user-attachments/files/20844978/public.xlsx)


For instance you can find result for mv3 xnnq_q8 S22 Ultra android 14:
```

Latency Stability Analysis: table10 (Primary)
================================================================================
Model: mv3(xnnpack_q8)
Device: Samsung Galaxy S22 Ultra 5G (private)(Android 14)

Dataset Overview:
  - Number of samples: 88
  - Date range: 2025-04-29 09:48:57+00:00 to 2025-05-13 21:08:36+00:00

Central Tendency Metrics:
  - Mean latency: 2.91 ms
  - Median latency (P50): 2.54 ms
  - Mean trimmed latency: 2.41 ms
  - Median trimmed latency: 2.15 ms

Dispersion Metrics:
  - Standard deviation: 1.14 ms
  - Coefficient of variation (CV): 39.08%
  - Interquartile range (IQR): 0.82 ms
  - Trimmed standard deviation: 0.76 ms
  - Trimmed coefficient of variation: 31.60%

Percentile Metrics:
  - P50 (median): 2.54 ms
  - P90: 3.88 ms
  - P95: 4.60 ms
  - P99: 5.91 ms

Inter-Jitter Metrics (variability between runs):
  - Max/Min ratio: 5.6103
  - P99/P50 ratio: 2.3319
  - Mean rolling std (window=5): 0.79 ms

Intra-Jitter Metrics (variability within runs):
  - Mean trimming effect ratio: 15.37%
  - Max trimming effect ratio: 38.83%

Stability Assessment:
  - Overall stability score: 0.0/100
  - Overall stability rating: Poor

Interpretation:
  The benchmark shows poor stability (score: 0.0/100) with significant
  variation between runs (CV: 39.08%).
  Performance is unpredictable and may lead to inconsistent user experience.

  The significant difference between raw and trimmed means suggests
  considerable intra-run jitter (15.4%) with occasional outliers within benchmark runs.

  The max/min ratio of 5.61 indicates
  substantial performance differences between the best and worst runs.

  The P99/P50 ratio of 2.33 suggests
  occasional latency spikes that could affect tail latency sensitive applications.
```

---------

Signed-off-by: Yang Wang <[email protected]>

Author: yangw-dev

.ci/docker/requirements-ci.txt

@@ -28,3 +28,6 @@ matplotlib>=3.9.4
 myst-parser==0.18.1
 sphinx_design==0.4.1
 sphinx-copybutton==0.5.0
+
+# script unit test requirements
+yaspin==3.1.0

.ci/scripts/benchmark_tooling/README.md

@@ -0,0 +1,172 @@
+# Executorch Benchmark Tooling
+
+A  library providing tools for fetching, processing, and analyzing ExecutorchBenchmark data from the HUD Open API. This tooling helps compare performance metrics between private and public devices with identical settings.
+
+## Table of Contents
+
+- [Overview](#overview)
+- [Installation](#installation)
+- [Tools](#tools)
+  - [get_benchmark_analysis_data.py](#get_benchmark_analysis_datapy)
+    - [Quick Start](#quick-start)
+    - [Command Line Options](#command-line-options)
+    - [Example Usage](#example-usage)
+    - [Working with Output Files](#working-with-output-files-csv-and-excel)
+    - [Python API Usage](#python-api-usage)
+- [Running Unit Tests](#running-unit-tests)
+
+## Overview
+
+The Executorch Benchmark Tooling provides a suite of utilities designed to:
+
+- Fetch benchmark data from HUD Open API for specified time ranges
+- Clean and process data by filtering out failures
+- Compare metrics between private and public devices with matching configurations
+- Generate analysis reports in various formats (CSV, Excel, JSON)
+- Support filtering by device pools, backends, and models
+
+This tooling is particularly useful for performance analysis, regression testing, and cross-device comparisons.
+
+## Installation
+
+Install dependencies:
+
+```bash
+pip install -r requirements.txt
+```
+
+## Tools
+
+### get_benchmark_analysis_data.py
+
+This script is mainly used to generate analysis data comparing private devices with public devices using the same settings.
+
+It fetches benchmark data from HUD Open API for a specified time range, cleans the data by removing entries with FAILURE indicators, and retrieves all private device metrics along with equivalent public device metrics based on matching [model, backend, device_pool_names, arch] configurations. Users can filter the data by specifying private device_pool_names, backends, and models.
+
+#### Quick Start
+
+```bash
+# generate excel sheets for all private devices with public devices using the same settings
+python3 .ci/scripts/benchmark_tooling/get_benchmark_analysis_data.py \
+  --startTime "2025-06-11T00:00:00" \
+  --endTime "2025-06-17T18:00:00" \
+  --outputType "excel"
+
+# generate the benchmark stability analysis
+python3 .ci/scripts/benchmark_tooling/analyze_benchmark_stability.py \
+--primary-file private.xlsx \
+--reference-file public.xlsx
+```
+
+#### Command Line Options
+
+##### Basic Options:
+- `--startTime`: Start time in ISO format (e.g., "2025-06-11T00:00:00") (required)
+- `--endTime`: End time in ISO format (e.g., "2025-06-17T18:00:00") (required)
+- `--env`: Choose environment ("local" or "prod", default: "prod")
+- `--no-silent`: Show processing logs (default: only show results & minimum logging)
+
+##### Output Options:
+- `--outputType`: Choose output format (default: "print")
+  - `print`: Display results in console
+  - `json`: Generate JSON file
+  - `df`: Display results in DataFrame format: `{'private': List[{'groupInfo':Dict,'df': DF},...],'public':List[{'groupInfo':Dict,'df': DF}]`
+  - `excel`: Generate Excel files with multiple sheets, the field in first row and first column contains the JSON string of the raw metadata
+  - `csv`: Generate CSV files in separate folders, the field in first row and first column contains the JSON string of the raw metadata
+- `--outputDir`: Directory to save output files (default: current directory)
+
+##### Filtering Options:
+
+- `--device-pools`: Filter by private device pool names (e.g., "samsung-galaxy-s22-5g", "samsung-galaxy-s22plus-5g")
+- `--backends`: Filter by specific backend names (e.g.,"xnnpack_q8")
+- `--models`: Filter by specific model names (e.g., "mv3", "meta-llama-llama-3.2-1b-instruct-qlora-int4-eo8")
+
+#### Example Usage
+
+Filter by multiple private device pools and models:
+```bash
+# This fetches all private table data for models 'llama-3.2-1B' and 'mv3'
+python3 get_benchmark_analysis_data.py \
+  --startTime "2025-06-01T00:00:00" \
+  --endTime "2025-06-11T00:00:00" \
+  --device-pools 'apple_iphone_15_private' 'samsung_s22_private' \
+  --models 'meta-llama/Llama-3.2-1B-Instruct-SpinQuant_INT4_EO8' 'mv3'
+```
+
+Filter by specific device pool and models:
+```bash
+# This fetches all private iPhone table data for models 'llama-3.2-1B' and 'mv3',
+# and associated public iPhone data
+python3 get_benchmark_analysis_data.py \
+  --startTime "2025-06-01T00:00:00" \
+  --endTime "2025-06-11T00:00:00" \
+  --device-pools 'apple_iphone_15_private' \
+  --models 'meta-llama/Llama-3.2-1B-Instruct-SpinQuant_INT4_EO8' 'mv3'
+```
+
+#### Working with Output Files CSV and Excel
+
+You can use methods in `common.py` to convert the file data back to DataFrame format. These methods read the first row in CSV/Excel files and return results with the format `list of {"groupInfo":DICT, "df":df.Dataframe{}}`.
+
+```python
+import logging
+logging.basicConfig(level=logging.INFO)
+from .ci.scripts.benchmark_tooling.common import read_all_csv_with_metadata, read_excel_with_json_header
+
+# For CSV files (assuming the 'private' folder is in the current directory)
+folder_path = './private'
+res = read_all_csv_with_metadata(folder_path)
+logging.info(res)
+
+# For Excel files (assuming the Excel file is in the current directory)
+file_path = "./private.xlsx"
+res = read_excel_with_json_header(file_path)
+logging.info(res)
+```
+
+#### Python API Usage
+
+To use the benchmark fetcher in your own scripts:
+
+```python
+from .ci.scripts.benchmark_tooling.get_benchmark_analysis_data import ExecutorchBenchmarkFetcher
+
+# Initialize the fetcher
+fetcher = ExecutorchBenchmarkFetcher(env="prod", disable_logging=False)
+
+# Fetch data for a specific time range
+fetcher.run(
+    start_time="2025-06-11T00:00:00",
+    end_time="2025-06-17T18:00:00"
+)
+
+# Get results in different formats
+# As DataFrames
+df_results = fetcher.to_df()
+
+# Export to Excel
+fetcher.to_excel(output_dir="./results")
+
+# Export to CSV
+fetcher.to_csv(output_dir="./results")
+
+# Export to JSON
+json_path = fetcher.to_json(output_dir="./results")
+
+# Get raw dictionary results
+dict_results = fetcher.to_dict()
+
+# Use the output_data method for flexible output
+results = fetcher.output_data(output_type="excel", output_dir="./results")
+```
+
+## Running Unit Tests
+
+The benchmark tooling includes unit tests to ensure functionality.
+
+### Using pytest for unit tests
+
+```bash
+# From the executorch root directory
+pytest -c /dev/null .ci/scripts/tests/test_get_benchmark_analysis_data.py
+```